Targeted therapies have the potential to improve cancer care by providing personalized treatment strategies based on genetic drivers. BRAF (~50%) and NRAS (~20%) have been identified as oncogeneic drivers in melanoma. Targeting mutant BRAF with specific inhibitors has demonstrated significant clinical impact resulting in recent FDA approval of several agents (vemurafenib and dabrafenib). Since NRAS has proven difficult to target pharmacologically, much effort has been directed at disrupting well defined downstream signaling pathways such as MAPK and PI3K signaling. Despite these efforts, targeting either or both of these pathways in mutant NRAS tumors has not been successful clinically, leaving few treatment options for a significant number of patients. Therefore, additional therapeutic targets need to be identified. We have recently developed a novel mouse model of melanoma that utilizes retroviral delivery of genes (e.g., NRAS) specifically to a small subset of melanocytes resulting in melanoma formation in the context of Cdkn2a loss (most common tumor suppressor loss observed in melanoma). While this model is well suited to address the role of a delivered gene(s) in tumor progression, it (like all other current models) lacks the versatility to eliminate any gene post tumor formation in an inducible manner. Crossing mice with different traditional or inducible knock outs would be time consuming and cost prohibitive and still primarily address tumor initiation rather than tumor maintenance which is more relevant to the human disease. Therefore, there is an urgent need to develop immune-competent pre-clinical models that allow for the specific deletion of any target for which pharmacological agents do not exist post tumor formation. To this end, we propose to advance our melanoma mouse model to incorporate inducible CRISPR (iCRISPR) technology to allow for rapid in vivo assessment of novel molecular targets. We have recently developed viral vectors capable of inducing NRAS driven melanomas in vivo with the capacity to delete any specific gene via tetracycline induction of CRISPR/Cas9. This versatile system enables rapid in vivo screening of any candidate genes' role in tumor maintenance. As proof of principle, in Aim 1 we will evaluate NRAS-directed iCRISPR in our NRAS melanoma model. Aims 2 and 3 will assess the role(s) of potential candidates implicated in NRAS tumor maintenance (Ccnd1 and Sos1).